Last year when 23andme had a $99 sale for DNA Day I jumped on the deal (a $400 savings!) and ordered a test kit. Since my results came in I’ve found that while they eventually may present some clear answers, for now the results have mainly provided a path to follow. And the path takes a LOT of reading.
With the help of others in the 23andme community and some user-run projects I’ve learned quite a bit, but I’ve only scratched the surface when it comes to understanding all of the science and history.
As of now I’ve got about 350 genetic matches on their Relative Finder, (mostly 5th cousins and higher) and am sharing genomes with several dozen. But it is not always obvious which side of my tree those cousins are from – I’m half Czech and half Irish and I have matches in quite a few countries, with some small clusters toward Germany, Ukraine, Russia and Scandinavia.
I’m guessing these matches are from further back in the tree my Czech side, but that’s all they are for now – guesses. (It is possible the Scandinavian goes back through my Irish side with all of the Vikings that came down for a visit.)
So, as any good techno-geek would do, I’ve tried to put crunch the data in various ways to look for patterns that might offer further clues. My first run at this was a simple chart of the Haplogroups of my matches in 23andme’s Relative Finder. My own maternal haplogroup is U5b1 and my paternal haplogroup is R1b1b2a1a2f.
Interestingly, in my maternal haplogroup matches I do not have any matches in my own U5b1 haplogroup and only 3 out of 195 in nearby U subclades.
My maternal matches are in 97 different haplogroups, the majority with one match each. (My paternal matches only span 27 different haplogroups. Also, on my paternal haplogroup side the large majority of my matches are in my own R1b1b2a1a2f, R1b1b2a1a2f2 and other nearby subclades. I’ll follow up to this blog post with my paternal chart.)When I first looked at my maternal haplogroup matches, I was just quizzical. Was there any reason for the higher matches in a small subset of groups? I posted the question on the 23andme community forum and some helpful members led me to the answer: The distribution of one’s matches should mimic the general population of where your ancestors came from.
This could be helpful if you are person who does not yet know your origin – you could determine the haplogroup frequency of your 23andme matches and then compare them to the country by country distribution listed at Eupedia’s Distribution of European mitochondrial DNA (mtDNA) haplogroups by region in percentage reference chart. (Say that 3 times fast.)
There is a caveat to this technique, though. The distribution of my own 23andme matches in several maternal haplogroups seems to be fairly close to Eupedia’s percentages for the Czech Republic and nearby countries. But, my matching percentages in my own U5b1 and the other U haplogroups is about half of the Eupedia numbers for those countries. If I didn’t have direct knowledge of my Czech ancestry that might lead me astray in my origin research.
For now I am assuming that it is the small sample size of my 23andme connections that is causing this difference, but will be interesting to watch as I get more data.
Next time: further up and further in!